Of the ~178,000 newly diagnosed cases of invasive breast cancer in 2008, the majority will be estrogen receptor- positive (ER+). Endocrine therapy produces a 26% proportional increase in overall survival. Nonetheless, almost 50% of all ER+ breast tumors will not respond de novo to such treatments, and many ER+ tumors that show an initial response will ultimately recur. Thus, resistance to endocrine therapy remains a significant clinical problem, and advanced ER+ breast cancer is still largely an incurable disease. A major limitation to progress in this field is the incomplete understanding of how cellular signaling to regulate cell fate is differentially affected by estrogens, antiestrogens, and aromatase inhibitors in sensitive and resistant breast cancers. Published and unpublished studies from our laboratory show that, in endocrine resistant cells, upregulation of the unconventionally spliced X-Box binding protein-1, (XBP1(S)), induction of the unfolded protein response (UPR), and concurrent signaling by RELA-driven NFB, may converge on BCL2 family members to regulate cell fate. Other key integrated signaling events likely also affect this process, which may represent a novel XBP1/endocrine-regulated signaling through the UPR that is specific to breast cancer. While the outcome of this signaling clearly affects the ability of endocrine manipulations to perturb mitochondrial function and initiate apoptosis, it is now evident that the balance between prosurvival and prodeath autophagy also plays an integral role. In this revised application (-A2), we now focus more closely on the role of XBP1/NFB (RELA) signaling in the context of the UPR, in conferring resistance to endocrine therapies in breast cancer. A better understanding of this cell signaling will lead to the discovery of novel therapeutic targets for improving response to endocrine therapies. We hypothesize that XBP1(S) and RELA are key regulators of endocrine responsiveness, and that their signaling through the UPR converge on BCL2 family members and also independently regulate cell survival. This convergence may reflect integration of signaling, resulting in an increase in prosurvival autophagy in endocrine resistant cells. Since XBP1, RELA, and their key functional mediators are associated with cell survival, we also hypothesize that improved predictors of clinical endocrine responsiveness and breast cancer prognosis can be built using unique combinations of these genes. We further hypothesize that innovative in silico modeling of breast cancer data from cell lines and patients will generate gene networks that can explain how cell survival signaling flows differentially in endocrine sensitive and resistant breast cancer cells. The primary goals of this application are to (a) study how XBP1(S) and RELA mechanistically affect endocrine responsiveness, and (b) explore the association of XBP(1), RELA and their signaling with clinical outcome in breast cancer patients. These goals will be addressed in a translational study, comprising three integrated specific aims, and performed by a multidisciplinary team of investigators with expertise in cellular/molecular biology, medical oncology, biostatistics, bioinformatics, and computer engineering. Aim 1: We will determine whether RELA and XBP1 interact at the level of BCL2 family members to affect endocrine responsiveness and determine cell fate. Aim 2: We will study whether measuring individual or collective protein and/or mRNA expression of XBP1, RELA, and BCL2 have predictive (response to endocrine therapy) and/or prognostic power (outcome independent of treatment) in sporadic breast cancers. Aim 3: We will build and test interactive in silico models of how XBP1, RELA, ER, and BCL2 signal to affect breast cancer cell survival in the context of UPR and endocrine responsiveness.